Stabilized platform systems have been around for many years, and are used to isolate a payload carried by the platform from the movement of the structure that carries the platform. The structure may be a vehicle like an airplane, helicopter or automobile, or a relatively static structure which is still subject to some movement, such as a tall pole that may sway in the wind. There is virtually no limit to what may be carried as the payload of a stabilized platform system, and stabilized platform systems may be used in a variety of applications for payloads including, but not limited to, still photographic and video (including cinema) cameras, electro-optical and infra-red imaging devices, spectrometers, antennae, lasers, and even weapon systems. What distinguishes this category of stabilization technology from others is that the platform that carries the payload is being stabilized and steered in inertial space. U.S. Pat. No. 4,796,090 to Fraier provides a detailed description of the need for platform stabilization in long range, high resolution, surveillance systems combined with the benefit of reduced integration times.
Various technologies for compensating for the movement of the structure carrying a sensor payload are known, each with drawbacks and limitations.
One approach for image-capturing payloads such as camera systems is to try to digitally stabilize the image captured by the payload, rather than stabilizing the payload itself. U.S. Patent Application Publication No. 20120019660A1 in the name of Golan describes the use of sequential image analysis, digital windowing and pixel shifting techniques as a means of digitally stabilizing the image and then further computing camera maneuvering signals to steer a coarse pan/tilt gimbal system. U.S. Pat. No. 7,876,359 to VonFlotow describes a similar digital stabilization technique, and U.S. Pat. No. 6,720,994 to Grottodden et al. describes a technique for adjusting the sample time between individual lines of pixels on the detector array as the image is captured. The issue with these digital stabilization techniques is that nothing is done to compensate for the motion of the payload's line of sight during the integration time period of the pixels that make up the image. This may result in motion-based blur in the captured image.
Other approaches seek to actually stabilize the payload relative to the supporting structure by stabilizing the platform that carries the payload. Within this “platform stabilization system” category there are passive and active systems. One example of a passive stabilization system is the STEADICAM® system described in U.S. Pat. Nos. 4,017,168 and 4,156,512 to Brown and U.S. Pat. No. 5,435,515 to DiGiulio et al. Another passive system is described in U.S. Pat. No. 5,243,370 to Slater. However, most platform stabilization systems make use of servomotors, inertial sensors, and a control system to augment the inherent inertia of the platform and are thus termed active systems.
Platform stabilization systems were initially developed to mount navigation instruments on moving vehicles such as ships and aircraft. Gyro compasses and vertical gyros, such as taught by U.S. Pat. No. 2,551,069 to Strother et al., are early examples of platform stabilization systems. Eventually, photographic cameras were mounted on these stable platforms to remove the unwanted motion of the vehicles during the acquisition of the image, for example as taught by U.S. Pat. No. 2,490,628 to Issertedt, U.S. Pat. No. 2,523,267 to Aschenbrenner et al., U.S. Pat. No. 2,883,863 to Karsten et al., U.S. Pat. No. 3,060,824 to Brenner et al. and U.S. Pat. No. 3,775,656 to Romans. Motion picture cameras, however, required more than just stability during the image acquisition; they also needed smooth steering control between the images.
New isolation mounts, such as those taught by U.S. Pat. No. 2,506,095 to Mantz, were developed to allow the camera to be manually steered while attenuating some of the vehicle vibration. Fixed gyros were added to the cameras to further improve stability and smoothness of steering. The camera operator typically sat in the open doorway of a helicopter with the camera, attached to an isolation mount with fixed gyros adding stability, placed over one shoulder. The camera operator would carefully coordinate with the pilot to steer the camera. This obviously made it quite difficult to frame the subject of the movie shot and achieve visually pleasing camera control.
In the late 1960s, Westinghouse Canada developed the WESCAM® platform stabilization system to address these issues. This was the first commercially available gyro stabilized, remotely steered camera system and is the subject of U.S. Pat. No. 3,638,502 to Leavitt et al. This type of stabilization technology relies on the angular momentum generated in three orthogonal, large mechanical rate gyroscopes (gimbaled flywheels) to augment the natural inertia of the camera platform. This artificial mass or synthetic inertia is used passively to maintain a slightly pendulous stable platform, with the payload (a camera) being steered relative to that stabilized platform. An active servo system then uses the angular rates measured by the precession of the gyros to cancel any disturbances using servomotors. A dome enclosure keeps the wind and weather out and an internal passive vibration isolation system minimizes the vibration input to the system.
The prior art for active platform stabilization technology can be classified into four general types or “generations”: gyro stabilized systems (first generation), classical active gimbal systems (second generation), limited travel—active follow-up systems (third generation) and unconstrained actuator—active follow-up systems (fourth generation). Within each generation there may be subtle differences in the implementation methods and advantages, however, the basic techniques are the same. The original WESCAM® platform stabilization system technology described in U.S. Pat. No. 3,638,502 is classified as first generation platform stabilization technology. It was further refined and a vertically slaved window was added, as described in U.S. Pat. No. 4,821,043 to Leavitt, to improve the optical performance of the system. Other first generation platform stabilization systems are described in U.S. Pat. No. 4,989,466 to Goodman and U.S. Pat. Nos. 5,184,521 and 5,995,758 to Tyler. While the first generation platform stabilization systems achieved significant stability, they suffered from poor steering bandwidth, which made them incompatible with video-trackers and required a highly skilled operator to compensate for this poor steering performance.
A second generation of active platform stabilization technology was developed to address the poor steering performance of the early first generation platform stabilization systems. These second generation platform stabilization systems, referred to as “classical active gimbal systems”, interpose a plurality of gimbals between the structure and the platform and close rate loops directly about each gimbal axis. Inertial rate sensors, such as small mechanical sensing gyros, are used to sense angular rates of the platform relative to inertial space. These rates are summed with the steering commands to stabilize and steer each axis. U.S. Pat. No. 3,986,092 to Tijsma et al., U.S. Pat. No. 5,868,031 to Kokush et al., U.S. Pat. No. 6,396,235 to Ellington et al., U.S. Pat. No. 7,000,883 to Mercadal et al., U.S. Pat. No. 8,100,591 to Chapman et al. and U.S. Pat. No. 8,564,699 to Bateman et al. are all examples of classical active gimbal systems. While each patent document describes subtly different methods and advantages, they all use a system of gimbals to support a platform, while closing rate loops directly about each gimbal axis using inertial rate sensors. The actuator can be either a direct-drive or a geared motor. The use of a geared actuator will increase coupling forces substantially, introduce backlash, and limit the steering bandwidth of the system. The structure between each successive gimbal axis is subjected to the high frequency torques of the actuators. Compliance in this constraint structure will limit the bandwidth of the control system. For this reason, classical active gimbal systems are generally incapable of high bandwidth performance with large payloads. U.S. Pat. No. 6,198,452 to Beheler presents an alternate, non-orthogonal, gimbal geometry for a classical active gimbal system, and U.S. Pat. No. 6,609,037 to Bless et al. describes a control system for a classical gimbal system that uses rate feedback and feed-forward control loops combined with position feedback and feed-forward control loops for each axis to further improve the steering performance. The classical active gimbal system was improved by the addition of an independent outer gimbal in the form of a dome enclosure with a vertically slaved window as described in U.S. Pat. No. 4,821,043 noted above and a passive isolator interposed between the dome and the inner platform stabilization system. The friction from the large gimbal bearings and motor brushes, combined with the structural resonances of the gimbal constraint system, limited the achievable stabilization performance of this system.
In order to further improve platform stability over that achieved by classical active gimbal systems, a third generation of active platform stabilization system was developed. It uses a higher bandwidth, limited travel inner gimbal mounted on a passive isolator, which in turn is mounted on the final stage of a low bandwidth, large travel outer follow-up gimbal system. As such, this type of platform stabilization system is referred to as a “limited travel-active follow-up” system. The inner gimbal provides the high bandwidth stabilization and fine steering performance, while the outer gimbal provides the coarse steering over a large field of regard. The inner gimbal uses high performance, direct drive actuators and the outer gimbal uses geared actuators. The high frequency torques are, however, still applied through the inner gimbals' constraining structure, but the inner gimbals' bearings are much smaller and the motors are typically brushless. While with smaller payloads, and with the use of fibre-optic gyros, the stabilization performance of this type of inner/outer gimbal system is satisfactory, with large payloads the compliance of the large gimbal ring structure limits the bandwidth of the stabilization system. U.S. Patent Application Publication No. 2010/0171377A1 in the name of Aicher et al. and U.S. Pat. No. 8,385,065 to Weaver et al. are recent examples of “limited travel—active follow-up” platform stabilization systems.
To address the bandwidth limitations caused by the structural resonances of the constraint system in the “limited travel—active follow-up” platform stabilization system, a fourth generation of active platform stabilization system was developed. This type of system, referred to herein as an “unconstrained actuator—active follow-up” system, avoids the bandwidth limitation of the “limited travel—active follow-up” system by using a process of torquing across the constraining structure instead of through it. The high frequency torques are applied directly from the outer gimbal to the platform. Combined with a high performance fibre-optic-gyro-based inertial measurement unit, this system raised the steering bandwidth significantly while maintaining stability. Examples of “limited travel—active follow-up” platform stabilization systems are described in U.S. Pat. Nos. 4,033,541 and 4,498,038 to Malueg, U.S. Pat. No. 4,828,376 to Padera, U.S. Pat. No. 5,368,271 to Kiunke et al., U.S. Pat. No. 5,897,223 to Tritchew et al., U.S. Pat. No. 6,196,514 to Kienholz, U.S. Pat. No. 6,263,160 to Lewis, U.S. Pat. Nos. 6,454,229 and 6,484,978 to Voigt et al. and U.S. Pat. No. 6,849,980 to Voigt et al. While each patent describes subtly different methods and advantages, they all:                use a system of intervening gimbals to support a platform on a support frame, while the gimbals constrain the platform's motion to limited rotation in three axes;        use an array of voice coil actuators which are configured to apply torques across, rather than through, the gimbal constraint system (sometimes across the gimbal and the isolator array in series); and        use an array of angular, inertial sensors to drive the voice coil motors to stabilize and steer the platform and thereby control the payload's line of sight.        
An alternate, non-orthogonal, inner gimbal configuration is presented in U.S. Pat. No. 4,733,839 to Gehris. The limited space available between the shells around the pivots suggests its intended use as either a “free gimbal”, missile seeker head, or unconstrained actuator—active follow-up platform stabilization system.
The primary problems with the current state of the art in active platform stabilization technology are cost, complexity, and reliability. The complex mechanical gimbal systems of the existing technologies are dominated by recurring costs. These include tight machining tolerances for bearing interfaces, the need for complex inspection and testing, precise alignment and preload of gimbal bearings during assembly, and ongoing inspection and maintenance.